© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 15428
    Prev
    Next
    INDEX
    Explanations

    terms related to the difficulty or ease of understanding concepts or data

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Juliushanhanhan/llama-3-8b-it-res/blocks.25.hook_resid_post
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    gated
    Context Size
    1,024
    Dataset
    Juliushanhanhan/openwebtext-1b-llama3-tokenized-cxt-1024
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    elo
    -0.17
    éIJ
    -0.17
    esson
    -0.16
    oine
    -0.15
    done
    -0.15
    compat
    -0.14
    uuid
    -0.14
    obra
    -0.14
    ÑĤÑĥÑĢа
    -0.14
    okt
    -0.13
    POSITIVE LOGITS
     to
    0.51
    to
    0.30
    _to
    0.30
    -to
    0.29
    	to
    0.28
     zu
    0.23
    ãĤĴ
    0.23
     To
    0.21
     Ø¥ÙĦÙī
    0.21
    To
    0.21
    Activations Density 0.039%

    No Known Activations